Suzaku, Courtesy JAXA

X-ray Imaging Spectrometer (XIS)
CCD Performance Monitoring

This page is intended to track the performance of the XIS CCDs independent of the time-dependent response applied by the standard calibration pipeline. As such, calibration corrections have not been applied in the plots displayed below. Please take care when comparing to other XIS data, and please read these caveats.


Calibration Source Event Monitoring

The Fe55 calibration sources illuminate the far corners of quads A and D of each detector. These figure tracks measurements of all grade 02346 events in those chip regions for each detector. Black points show SCI-off data, blue points show SCI-on data, and red points show data taken at non-standard CCD temperatures (as identified in the temperature history file). The red points are excluded from the fitting. Monthly accumulations of calibration source data are also available as spectra.

Figure 1 shows shows the mean normalized energy of the Mn K alpha line, assuming a peak energy of 5.894 keV (top); the FWHM in eV of the line (middle); and the count rate in the line (bottom). Also available is the text data file from which the figure was produced.

Updated 2014-06-12
Download PDF - data file

Figure 2 shows shows the mean normalized energy of the Mn K beta line, assuming a peak energy of 6.490 keV (top); the FWHM in eV of the line (middle); and the count rate in the line (bottom). Also available is the text data file from which the figure was produced.

Updated 2014-06-12
Download PDF - data file

Current gain and FWHM change measurements at Mn K alpha (5.9 keV) and Mn K beta (6.5 keV):

change in gain (%/year)
SCI offSCI on
Mn K alphaMn K betaMn K alphaMn K beta
XIS0-1.805 +/- 0.003-1.755 +/- 0.008-0.391 +/- 0.001-0.385 +/- 0.002
XIS1-1.967 +/- 0.004-1.870 +/- 0.009-1.014 +/- 0.001-1.008 +/- 0.004
XIS2-1.732 +/- 0.002-1.757 +/- 0.021-0.325 +/- 0.037a 0.011 +/- 0.657a
XIS3-1.783 +/- 0.004-1.772 +/- 0.008-0.360 +/- 0.001-0.358 +/- 0.002
change in FWHM (eV/year)
SCI offSCI on
Mn K alphaMn K betaMn K alphaMn K beta
XIS067.0 +/- 1.148.8 +/- 1.914.2 +/- 0.420.6 +/- 1.1
XIS166.9 +/- 0.956.2 +/- 2.628.3 +/- 0.542.2 +/- 2.3
XIS259.2 +/- 0.947.1 +/- 5.4.... +/- ...a.... +/- ...a
XIS366.9 +/- 1.452.6 +/- 2.214.5 +/- 0.217.5 +/- 1.0
a - XIS2 ceased operations shortly after routine SCI-on observations were begun,
so the gain and FWHM trends are poorly constrained for this mode due to insufficient data.


The locations of the cal source extraction regions are identified below. The image shows a single CCD field (XIS2) in ACTX,ACTY coordinates, with cal sources illuminating the upper left and upper right corners. The coordinates in the table give the inside corner of the extraction boxes for each chip. The outside corners are (x,y) = (0,1024) for quad A and (1024,1024) for quad D.

     
  x1,y1 x2,y2
XIS0 200,750 824,750
XIS1 250,800 824,800
XIS2 250,800 850,800
XIS3 200,800 800,750

Pixel-based Monitoring

This figure shows the corner pixel histogram mean (top) and sigma (middle) for all 3x3 events, and the number of hotpixels in the eventlist. The BI chip (XIS1) has different trends from the FI chips because of the difference in event thresholds, which are 100 ADU (FIs) and 20 ADU (BI). Corner pixel information is gathered from the cal source regions as defined above. Hot pixels are extracted from the full CCD field and require that at least 10% of the frames have an event in that pixel.

Updated 2014-06-12
Download PDF - data file


CTI Monitoring

This figure shows the ratio between top split events and bottom split events (top), and the ratio between right split events and left split events (bottom). These can be used to track changes in parallel and serial CTI, respectively. Grades shown are ACIS 8-bit grade codes.

Updated 2014-06-12
Download PDF - data file

This figure shows the vertical trailing pixel histogram centroid (top) and 1-sigma width (bottom). These can be used to track changes in CTI.

Updated 2014-06-12
Download PDF - data file

The trends in the trailing pixel centroid:

change in trailing pixel centroid
SCI offSCI on
ADU/year%/yearADU/year%/year
XIS01.61 +/-0.020.47 1.73 +/-0.010.39
XIS11.40 +/-0.020.13 1.78 +/-0.010.15
XIS21.40 +/-0.020.37 1.58 +/-0.71a0.34 a
XIS31.47 +/-0.020.37 1.74 +/-0.010.39
a - XIS2 ceased operations shortly after routine SCI-on observations were begun,
so the gain and FWHM trends are poorly constrained for this mode due to insufficient data.



Important Caveats

The plots on this page are intended to characterize the XIS CCD performance, not the quality of processing or of the response calibration. The data used are raw FFF files, processed only by the mk1stfits routine to convert from RPT to FITS format. Processing such as event grading, filtering, and PHA to energy conversion are performed in a different way than the standard ISAS/GSFC Suzaku pipeline, with most calibration corrections (e.g. CTI correction) not performed at all. Therefore, please take care when comparing these trends to XIS data from other sources.

The basic steps of processing to produce these trend plots are:

  1. copy fff files from astro.isas.jaxa.jp:/usr/local/astroe/data/xis/yyyymmdd/
  2. collect all full frame 3x3 and 5x5 data for a given day and strip out only the time, ccdx, ccdy, segment, and nine central pixel pulse heights.
  3. determine the asca grade for each event (using a split threshold that is constant for each device: 13adu for the FIs, 7adu for the BIs. it does NOT vary as a function of energy.)
  4. determine the summed pulse height by adding the edge pixels above split threshold for all g2346 events. also add the attached corner pixel for the g6 events.
  5. convert to energy using only the slope of the gain as determined by cal source data from just after launch.
  6. make an energy spectrum from only the g02346 events from the cal source region
  7. do a simple three parameter (height, location, and width) gaussian fit around the expected location of the main Kalpha and Kbeta peaks
  8. plot each of the line locations divided by the expected location as a function of date for each of the main peaks.


MKI     
MIT
Last updated: Thu Jun 19 13:47:08 EDT 2014
email: milleric@mit.edu